The use of color in the digital environment has created problems for color printers trying to produce satisfactory results. One problem facing color printers stems from the proliferation of desktop publishing software programs or applications.
The problem with these desktop publishing systems is that the systems allow the user to combine different types of objects into a composite document. For example, a user can combine photographic images, text, and business graphics (charts) into a single document wherein these images may be either color or black/white.
To achieve satisfactory results each of these objects need to processed differently so that the high quality document can be produced. For example, let assume that a digital color system is trying to render a composite document with a photographic image and a business graphic. In order to achieve high quality rendering of a photographic image, the color system may have to skew the color attributes in a certain way, but this skewing may cause the business graphics in the same composite document to appear washed out. On the other hand, if the color printing system is skewed to ensure saturated vivid colors for the business graphics, the photographic image in the composite document may lose its life-like appearance.
To resolve this problem, object oriented rendering systems have been developed. In such conventional systems, the objects which make up a composite document are rendered (processed) uniquely. In other words, a photographic object or object or bitmap will be processed one way, while business graphics will be processed another way. In this way, an object can be rendered to optimize its quality.
Furthermore, the neutral rendering requirements for different object types may vary. More specifically, neutral rendering refers to the appearance of black, greys, and white on a printed page or output device. This rendering may be understood in terms of process (multi-color) or true (single component) neutrals. Table 1 below provides an illustration as to how neutral rendering may vary from object to object.
TABLE 1OBJECTBLACKSGREYSWHITESBitmapProcess BlackProcess blackDoes not haveto be PaperWhiteGraphicSingleSinglePaper WhiteComponentComponentBlackBlackTextSingleSinglePaper WhiteComponentComponentBlackBlackPhotographicProcess BlackProcess BlackDoes not haveto be PaperWhite
As illustrated in Table 1, process blacks and greys may be acceptable for bitmap objects since true blacks or greys may introduce unwanted gloss differences. On the other hand, true blacks or greys are desired for black lines within a graphic object so as to reduce misregistration and imperfect color balance artifacts. Thus, an object may need to have its neutral image rendered in a manner different from the rendering of the non-neutral image. These diverse requirements cause problems when rendering objects in a composite image.
One example of this problem is the presence of an appreciable gloss differential when black text is to be rendered within a business graphic object. One conventional solution is to have the object rendered utilizing all of the primary colors resulting in the black text being rendered using process black (cyan, magenta, and yellow). By rendering the text with process black, the text will have the appearance of having a high gloss which is not necessarily desirable from a user's standpoint. Most user prefer a low gloss for text which cannot be realized with process black. Another possible conventional solution, from the standpoint of the obtaining neutrals, is the utilization of a 100% under color removal (UCR) strategy. However, this strategy may reduce the chroma of shadow regions and suffer from other color fidelity problems.
Another example of this problem is white point shifts caused by expert color control color adjustments, which allow for user control of various color attributes, such as lightness, contrast, chroma, and color balance. White point shifts are desirable for bitmap objects since these types of shifts occur in natural scenes; however, in graphic and text objects, it is desirable to have the expert color control adjustment leave the white points unaltered and unshifted since white points shifting in graphic and text objects create undesirable artifacts.
Not only do various objects require different neutral rendering processes, but the neutral items within an object may each require unique rendering processes. For example, an object may be classified as a graphic object. Moreover, this graphic object type may include a black line and/or fill. Although black lines and fills share many common rendering characteristics as members of the graphic object, black lines and fills differ with respect to neutral rendering. More specifically, black lines should be rendered as single component black rather than process black to reduce misregistration and imperfect color balance artifacts. On the other hand, black fills should be rendered as process black rather than single component black in order to reduce differential gloss effects that may be evident in color and black sweeps and to reduce black fill dropouts due to gloss differences with filled color areas.
Therefore, it is desirable to utilize an object oriented processing and rendering system which allows for object oriented neutral rendering and provide a framework for features that effect neutral rendering. Moreover, it is desirable to have an object oriented neutral processing and rendering system which optimizes the rendering of neutral colors based upon partitioning of objects and of types which can be further optimized by partitioning each object type into object sub-types. In other words, it is desirable to have a processing and rendering system which allows for common rendering characteristics to be identified by the object type and rendering fine tuning to be performed based upon the object sub-type.